Here's one of the abovementioned articles on the matter:

Rational Irrationality

Alfred Mele

One way to improve our understanding of some things is to figure out why they malfunction or break down. For the most part, that has been my approach in previous work on rationality. I have been especially concerned with breakdowns in belief and action. Here I concentrate on belief.

Thomas Gilovich reports that a survey of people who teach at universities “found that 94% thought they were better at their jobs than their average colleague.” Are university teachers exceptional in this regard? Seemingly not. Gilovich also reports that “A survey of one million high school seniors found that […] all students thought they were above average” in their “ability to get along with others” and “25% thought they were in the top 1%.” Obviously, if the respondents are being sincere, a great many of them are overestimating themselves.

Is there anything irrational in Alan's believing that he is more congenial than 99% of his peers when, in fact, over 20% of his peers are more congenial than he? According to one conception of rational belief – call it the evidential conception – it is a truism that rational belief is appropriately based on at least the readily available evidence. If Alan's belief about his congeniality is not so based, it is not a rational belief from the perspective of this conception.

Statistics of the kind I mentioned indicate that we have a tendency to believe propositions that we want to be true even when an impartial review of readily available data would indicate that they probably are false. A plausible hypothesis about that tendency is that our wanting something to be true sometimes biases what we believe. To the extent that a belief is biased in that way, it is not appropriately based on evidence.

In an interesting study by psychologist Ziva Kunda, undergraduate subjects read an article alleging that women are endangered by caffeine and are strongly advised to avoid caffeine in any form, that the major danger is a disease closely associated with breast cancer, and that caffeine causes the disease by increasing the concentration of a substance called “cAMP” in the breast. Because the article does not personally threaten men, they were used as a control group. Subjects were asked to indicate, on a 6-point scale, how convinced they were of the connection between caffeine and the disease and the connection between caffeine and cAMP. Women who drank a lot of coffee (heavy consumers) were significantly less convinced of the connections than were women who drank little coffee. The men were considerably more convinced than the female heavy consumers. And there was a much smaller difference in conviction between men who drank a lot of coffee and men who drank little (the heavy consumers were slightly more convinced of the connections).

Given that all subjects were exposed to the same information and assuming that only the female heavy consumers were personally threatened by it, a plausible hypothesis is that their lower level of conviction is motivated in some way by their wanting it to be false that their coffee drinking has endangered their health. Indeed, in a study in which the reported hazards of caffeine use were relatively modest, female heavy consumers were no less convinced by the evidence than were female light consumers. Along with the lesser threat, there is less motivation for skepticism about the evidence.

How do the female heavy consumers come to be less convinced than the others? One testable possibility is that because they find the connections at issue personally threatening, these women (or some of them) are motivated to take a hyper-critical stance toward the article, looking much harder than other subjects for reasons to be skeptical about its merits. Another is that, owing to the threatening nature of the article, they (or some of them) read it less carefully than the others do, thereby enabling themselves to be less impressed by it. In either case, to the extent that the women's beliefs are shaped by what they want to be true rather than being appropriately based on the evidence, these beliefs are irrational, according to the evidential conception of rational belief.

If motivationally biased beliefs of the kind I have been discussing are irrational, should we try to avoid all of them? Not necessarily. There is a phenomenon called “depressive realism”. Depressed people tend to be significantly more accurate about their positive and negative attributes than do people who are not depressed. Whether depression is a cause of the accuracy or the accuracy is a cause of the depression is an open question. But should you want to cause yourself to be depressed so that you can be more accurate about yourself or work hard to be more accurate about yourself at the risk of causing yourself to be depressed?

On many topics, a bit of self-overestimation is harmless and even useful. One potential effect of your overestimating your congeniality is your being more outgoing than you would otherwise be, making more friends, and having a better time socially. That sounds good, not bad. Similarly, a potential effect of my overestimating my teaching prowess is my feeling more at ease in the classroom than I otherwise would, with beneficial consequences. Of course, if I think I'm doing a superb job when in fact I'm boring my students to tears, I have little incentive to improve – a bad thing. But there is a difference between modest and massive self-overestimation.

If it is not true that we should try to rid ourselves of all of our motivationally biased beliefs about ourselves, are they really all irrational? Well, it might be that some irrationality is permissible and even helpful, and because it is, we have no obligation to root it out. Also, in the grand schemes of our lives, with all we have to do, is it really worth the time and trouble to bring into conformity with the facts our beliefs about how intelligent, attractive, likeable, nice, amusing, and so on we are – unless we seriously misestimate these things and that causes us significant problems? Surely, most of us have better things to do with our time.

Would it be irrational, then, to try to rid ourselves of as many of our irrational beliefs about ourselves as we can? If so, are we stuck with a paradox? Here we come to practical rationality and irrationality. According to one conception of practical rationality, it is rational to do what you, on the basis of good reasons, believe it best on the whole to do. (It may be rational to do other things too, when you have no belief about what it would be best on the whole to do in your circumstances.) I have suggested, in effect, that, for many of us anyway, including me, it would be best on the whole to refrain from trying to eradicate all of our irrational beliefs. I believe this – and for good reason, I think. Suppose my reasons are, in fact, good. Then, according to the conception of rationality now in play, it is rational of me to refrain from undertaking this project.

There is no paradox in this. Beliefs and actions are two very different kinds of thing. Because that is so, what justifies the judgement that certain of your beliefs are irrational might not also justify the judgement that it would be rational for you to act in such a way as to eradicate those beliefs.

Suppose you learn of a kind of psycho-surgery that enables people to bring all of their beliefs about their positive and negative attributes into line with the facts. Suppose you also learn that only this psycho-surgery would eliminate all of your biased beliefs about yourself, that it is very expensive, and that it would probably cut ten years off your life. Would it be rational for you to sign up for the surgery? Obviously not.

Suggested reading
How We Know What Isn't So , T Gilovich (Macmillan)
Irrationality , A Mele (Oxford University Press)
Self-Deception Unmasked , A Mele (Oxford University Press)

Alfred Mele is William H and Lucyle T Werkmeister Professor of Philosophy at Florida State University